Companies struggle with the long-term storage of low-touch data that, because of its nature, is realistically not a good fit for the high-priced, high-performance enterprise storage many companies rely on.
Most IT folks will acknowledge that stale, infrequently used and abandoned data causes them ongoing problems. This (usually) unstructured data is sometimes referred to as low-touch data. The issues associated with this data are many. For example, it can consume large amounts of high-priced enterprise storage causing IT to purchase more every year. It can extend the backup window so that employee productivity is potentially affected. And it will complicate and drive up the cost of eDiscovery.
The challenge with stale data is determining whether it should be retained for regulatory compliance, current or anticipated litigation, business reasons, or because the data is still useful to individual employees. In fact, the Compliance, Governance & Oversight Counsel (CGOC) published a survey stating that, on average, 1% of enterprise data is subject to litigation hold, 5% to regulatory retention and 25% had some business value. This leaves 69% of enterprise data possibly valueless. This potentially low-touch data includes expired data, aging data from current employees, and abandoned data from ex-employees. However, some percentage of this data is still valuable to the company and individual employees, so it’s important to ensure only truly valueless data is disposed of.
Another issue with the huge, unmanaged (and unindexed) volumes of stale data is finding specific content when you need it. Finding data when it’s not indexed or managed is a major cost for companies, especially during eDiscovery. To make the point, a CEO from my past, T. M. Ravi, CEO of Mimosa Systems, once stated, “It costs up to 500 times more to find and utilize information once, than to store it untouched for 20 years.”
T.M. Ravi’s insight highlights the point that leaving unindexed, unmanaged data “where it lies” is not an effective strategy, considering the majority of unstructured enterprise data rapidly ends up as abandoned data.
This is not to say this data is bad, just a costly challenge if not managed correctly. The obvious answer is to recognize and manage low-touch data in a repository specifically set up for it. A possible strategy includes directing employees to retain data on specific storage resources (not their local computers) for a period of time. As its last accessed date ages (for example 2 years), that data could be automatically indexed, identified and migrated to a centralized repository only accessible by IT or legal. This would ensure the low-touch data is managed, searchable and disposed of when appropriate.
In the end, it’s important to actually manage stale data — not ignore it.